Decoupling As(III) & As(V) Redox States For Surface Complexation

[awillou4]

Hello all,

I'm in the process of following example 9 to decouple As(3) and As(5) in my surface complexation model.  I'm doing so because it is likely that my laboratory arsenic solution is in redox disequalibrium, so I'm hoping to isolate As(3) for batch isotherm curve fitting purposes.  Example 9 in the PHREEQC manual doesn't appear to be a complexation example, so I'm trying to work out the necessary modifications for such a model.

My understanding is that I'll need to add in modified As(3) SOLUTION_MASTER_SPECIES and SOLUTION_SPECIES for all As(3) species.  This would look something like this for the wateq4f database:

SOLUTION_MASTER_SPECIES
    As_tri        As_tri+3         0     As_tri          74.922

SOLUTION_SPECIES
As_tri+3 = As_tri+3
        log_k   0.0

#As(3)  secondary master species   487
   H3As_triO4 + H2 = H3As_triO3 + H2O
   log_k   22.5
   delta_h   -117.480344   kJ

#H2AsO3-       478   
   H3As_triO3 = H2As_triO3- + H+
   log_k   -9.15
   delta_h   27.54   kJ

and so on until all 7 As SOLUTION_SPECIES in wateq4f are included.  I'm wondering if I should exclude the As second master species since it relates A(3) to As(5)?

I'm less certain about what comes next.  Since mine is a surface complexation model, I imagine I'll need to add in new surface species for my modified Arsenite.  This would look like this:

SURFACE_SPECIES
#   Arsenite
        Hfo_wOH + H3As_triO3 = Hfo_wH2As_triO3 + H2O
        log_k   5.41

After the new SURFACE_SPECIES is defined, do I need to add anything else like PHASES?  There are 17 phases which include As in wateq4f so I'd like to figure out if they're necessary before I delve into adding them.  Aside from PHASES, I think that would be everything; however I could certainly be missing something.

Any help is much appreciated!

[dlparkhurst]

You have the basic idea. If you are defining only As_tri, you would not include the reaction of As(5) to As(3).

If you want to use equilibrium with arsenic minerals, then you would need to define the ones needed with PHASES.

However, before you go too far, you can force As(3) to be the dominant phase in a couple ways without redefining the entire database.

(1) As long as the pe is sufficiently low, all of the arsenic will be As(3). You want it low enough to maintain As(3), but not so low as to cause other unwanted redox reactions, like generating H2 or CH4. So it depends in part on you system.

(2) Alternatively, and probably better, I think you should be able to redefine the As(5) -> As(3) reaction so that As(3) is heavily favored. Here I changed the K to 100. I think you should not get any As(5) and it would be equivalent to redefining As_tri. Check out the following:

SOLUTION_SPECIES
#As3  secondary master species   487
   H3AsO4 + H2 = H3AsO3 + H2O
   #log_k   22.5
        log_k   100.
   delta_h   -117.480344   kJ
SOLUTION
As 1
-pe 12 O2(g) 0
END

[awillou4]

Redefining the As(3)-> As(5) reaction was simple and did the trick.  My isotherm fits are looking about as good as these things get.  Thank you for your help!